In general, conscious perception of sensory stimuli depends on a pathway to neocortex, and in all modalities except for olfaction, this requires going through the thalamus. Careful anatomical studies have found that in olfaction, the pathway between olfactory cortex and prefrontal cortex is mostly direct, with only a small contingent of fibers going to mediodorsal thalamus. This fundamental difference in the structural organization of the olfactory system indicates a qualitatively different role for the thalamus in olfactory processing. The goal of this proposal is to elucidate that role. In Aim 1, I will use fMRI to characterize the nature of intentional modulation of activity in the thalamus. By systematically altering attention to different odors, I will test the hypothesis that distributed codes of odor quality are attention dependent in the mediodorsal nucleus of the thalamus, while they are attention independent in primary olfactory cortex. In Aim 2, I will use fMRI to Using hierarchical cluster analysis and a general linear model approach to data analysis, I will be able to distinguish between olfactory brain regions that are involved in basic sensory processing, accumulation of information and moment of recognition during odor valence decisions and quality recognition. In Aim 3, I will use fMRI to ask whether the mediodorsal thalamus integrates information from multiple modalities during olfactory decisions. During scanning, subjects will smell odors that are either presented alone or paired with stimuli from other modalities, enabling us to test for multimodal olfactory response profiles in the mediodorsal thalamus.